Machine Vision

cancel
Showing results for 
Search instead for 
Did you mean: 

Help with using the LearnCalibration.vi

Solved!
Go to solution

Hey Hummer,

 

I see you're right with the acceptable inputs. It is just when you are using an image that you need to have those data types:

 

"When learning a calibration template from an image of a grid of circles, the image type must be either U8 or I16. "

 

 Now, I'm not sure if this is because the software has trouble writing calibration data to anything else, or because the software has trouble reading the grid of circles from anything else. It, however, does seem to be the latter. 

 

Is there any reason in particular that you are wanting to use a SGL (grayscale)? It just seems to me that if it is possible to calibrate with an I16, that there would be no need to convert it to SGL, butyou might have a good reason for it, I'm just curious.

 

Something else that would probably work, but would need to be tried would be to calibrate with an I16, and then after calibrating the image, casting it to a SGL. The only thing that I am not positive about is whether or not you will lose calibration data on the image when its casted. 

Chris Van Horn
Applications Engineer
0 Kudos
Message 11 of 12
(918 Views)

We are working on astronomical images.  To process the images we take lots of averages and wind up with fractional results, thus the need for floating point.  I think we can use any blank image (all 0's in I16 format) to LearnCalibrate by providing the list of coordinates to both inputs...Then we can use the calibration data to correct the single precision image later.

 

I'll give that a try later and see how it works.

 

Thanks.

0 Kudos
Message 12 of 12
(916 Views)